CoarsenConf: Equivariant Coarsening with Aggregated Attention for Molecular Conformer Generation
Journal of Chemical Information and Modeling · 2025
DOI: 10.1021/acs.jcim.4c01001 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Chemical Information and Modeling, 2025. Verfasst von Danny Reidenbach und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Chemical Engineering, Department of Computer Science, University of California Berkeley, Berkeley, California 94720, United States. Hauptthemen laut MEDLINE: Molecular Conformation, Cheminformatics. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Laboratory Directed Research and Development, Google Cloud.
Abstract (Originalsprache)
Molecular conformer generation (MCG) is an important task in cheminformatics and drug discovery. The ability to efficiently generate low-energy 3D structures can avoid expensive quantum mechanical simulations, leading to accelerated virtual screenings and enhanced structural exploration. Several generative models have been developed for MCG, but many struggle to consistently produce high-quality conformers for meaningful downstream applications. To address these issues, we introduce CoarsenConf, which coarse-grains molecular graphs based on torsional angles and integrates them into an SE(3)-equivariant hierarchical variational autoencoder. Through equivariant coarse-graining, we aggregate the fine-grained atomic coordinates of subgraphs connected via rotatable bonds, creating a variable-length coarse-grained latent representation. Our model uses a novel aggregated attention mechanism to restore fine-grained coordinates from the coarse-grained latent representation, enabling efficient generation of accurate conformers. Furthermore, we evaluate the chemical and biochemical quality of our generated conformers on multiple downstream applications, including property prediction and large-scale oracle-based protein docking. Overall, CoarsenConf generates more accurate conformer ensembles compared to prior generative models.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- Department of Chemical Engineering, Department of Computer Science, University of California Berkeley, Berkeley, California 94720, United States.
- NVIDIA, Santa Clara, California 95051, United States.
ORCID-Profile: Danny Reidenbach
Förderung
- Laboratory Directed Research and Development
- Google Cloud
Zitieren
APA
Reidenbach, D., & Krishnapriyan, A. S. (2025). CoarsenConf: Equivariant Coarsening with Aggregated Attention for Molecular Conformer Generation. Journal of Chemical Information and Modeling. https://doi.org/10.1021/acs.jcim.4c01001
Vancouver
Reidenbach D, Krishnapriyan AS. CoarsenConf: Equivariant Coarsening with Aggregated Attention for Molecular Conformer Generation. Journal of Chemical Information and Modeling. 2025. doi:10.1021/acs.jcim.4c01001
BibTeX
@article{Reidenbach2025,
author = {Reidenbach, Danny and Krishnapriyan, Aditi S},
title = {CoarsenConf: Equivariant Coarsening with Aggregated Attention for Molecular Conformer Generation},
journal = {Journal of Chemical Information and Modeling},
year = {2025},
doi = {10.1021/acs.jcim.4c01001}
}
EUVIMED